Do LLM-assisted peer reviews violate bioRxiv copyright? Scholars clash on X
anshulkundaje · x · 2026-10-28
Stanford professor Anshul Kundaje and @WalentekLab clash over uploading preprints under review to frontier LLMs. WalentekLab argues LLMs can't guarantee no plagiarism, so uploading violates bioRxiv's copyright clause — and reviewers without time shouldn't take on reviews. Kundaje counters that this is almost surely false: bioRxiv authors retain copyright, choose their own license (including no-reuse options), and explicitly consent to text mining by researchers and search engines; if the uploader picks the no-training option and the company legally commits to it, no violation occurs. He adds that plagiarism by a frontier company claiming otherwise would be the actual copyright violation, and says he'll keep reviewing preprints openly with AI assistance, citing bugs AI found that he never would have. (Duplicate fragments of the same exchange in this batch are merged into this entry.)
More from Companies & People
- QNX expands from millions of vehicles to robotics and physical AI — pdamodaran · 2026-10-28
- Andreessen: AI gives founders managerial superpowers, ending Burnham's managerial society — robleclerc · 2026-10-28
- Researcher questions Dario: where's the evidence Anthropic's 'race to the top' is working? — adam_karvonen · 2026-10-28
- OpenAI x Oracle: ChatGPT gains Oracle apps and data, US customers can buy via Marketplace — OpenAIDevs · 2026-10-28
- Surge AI at Kaggle Bench Summit: why frontier models still struggle with work-like benchmarks — echen · 2026-10-28
- Harvard releases 6 free AI lectures covering prompt engineering and RAG — hey_abusiddik · 2026-10-28